R. CELIN et al.: PROPERTIES OF THE SIMULATED COARSE-GRAINED MICROSTRUCTURE ...
115–120
PROPERTIES OF THE SIMULATED COARSE-GRAINED
MICROSTRUCTURE OF QUENCHED AND TEMPERED
HIGH-STRENGTH STEEL
LASTNOSTI SIMULIRANE GROBOZRNATE MIKROSTRUKTURE
POBOLJ[ANEGA VISOKOTRDNEGA JEKLA
Roman Celin
1*
, Fevzi Kafexhiu
1
, Grega Klan~nik
2
, Jaka Burja
1
1
Institute of Metals and Technology, Lepi pot 11, 1000 Ljubljana, Slovenia
2
RCJ d.o.o., Cesta Franceta Pre{erna 61, 4270 Jesenice, Slovenia
Prejem rokopisa – received: 2020-07-28; sprejem za objavo – accepted for publication: 2020-10-05
doi:10.17222/mit.2020.144
Quenched and tempered high-strength steels have a favorable strength-to-weight ratio. These steels are usually welded with the
use of standard procedures. Due to the heat input during welding, steels undergo changes in the heat-affected zone and as a re-
sult an unfavorable microstructure and mechanical properties can occur. The coarse-grained heat-affected-zone microstructure
was studied with the use of simulated thermal cycles. They were applied to the experimental quenched and tempered steel. The
steel samples were heated to a peak temperature of 1350 °C and then cooled with different rates. A microstructure examination,
dilatometry analysis and hardness measurements of the steel samples were performed. The transformation start and finish tem-
peratures were determined using the dilatation vs. temperature data analysis. The investigation was performed to determine the
appropriate cooling time t8/5 interval for further investigation of the experimental steel.
Keywords: microstructure, coarse-grained, continuous cooling, hardness
Visokotrdna kaljena in popu{~ena jekla imajo ugodno razmerje med trdnostjo in te`o. Ta jekla se lahko varijo z uporabo
standardnih postopkov varjenja. Zaradi vnosa toplote med varjenjem se jeklo v toplotno vplivanem podro~ju spremeni, zaradi
~esar lahko pride do ne`elene mikrostrukture in slabih mehanskih lastnosti. Simuliacija varilnih toplotnih ciklov je bila
uporabljena za raziskavo grobo-zrnate mikrostrukture preizkusne {ar`e kaljenega in popu{~enega jekla. Vzorci jekla so bili
segreti do temperature 1350 °C in nato podvr`eni razli~nim hitrostim ohlajanja. Izvedena je bila analiza podatkov, zabele`enih z
dilatometrom, ter meritve trdote in preiskava mikrostrukture vzorcev. Za posamezne vzorce sta bili dolo~eni temperaturi za~etka
in konca transformacije pri ohlajanju. Rezultati izvedenih preiskav so osnova za dolo~itev primernega ~asa ohlajanja t8/5 za
nadaljne reziskave preizkusnega jekla.
Klju~ne besede: mikrostruktura, grobo zrno, kontinuirno ohlajevanje, trdota
1 INTRODUCTION
High-strength quenched and tempered steels are com-
monly referred to as Q&T steels. They are used to simul-
taneously reduce the weight and increase the load capac-
ity of load-bearing components (pressure vessels,
building columns, cranes, off-shore structures, etc.).
These steels are also used for the production of armored
steel plates.
1,2
The standard EN 10025-6 for quenched and tempered
steels allows variations of the chemical composition, but
with minimum requirements for the mechanical proper-
ties.
3
This allows steel manufacturers to use different
chemical compositions with a combination of appropri-
ate rolling parameters and heat treatment (quenching fol-
lowed by tempering) to achieve a minimum yield
strength of above 1000 MPa.
4,5
The final result is typi-
cally a microstructure of tempered martensite and
bainite.
Different arc-welding procedures such as shielded
metal arc welding (SMAW), gas metal arc welding
(GMAW), gas tungsten arc welding (GTAW) can be used
for joining quenched and tempered steels.
6
The properties of the weld metal and the micro-
structural development of the adjacent heat-affected zone
(HAZ) are to a large extent influenced by the welding
thermal cycle and the composition of the base material.
The influence of the welding thermal cycle can be ex-
pressed through the welding heat input, which can be de-
termined by the cooling time t
8/5
. The cooling time is de-
fined as the time for the temperature to drop from 800 °C
to 500 °C and is in an inverse relation to the cooling rate
(°C/s).
7
The longer the t
8/5
(higher heat input), the slower
the cooling rate is.
During welding the highest temperature, besides
weld molten metal, is in the coarse-grained heat affected
zone (CG HAZ) and can be above 1100 °C.
8
At that tem-
perature the initial tempered martensite microstructure of
the quenched and tempered steel undergoes an austenite
transformation and the grains start to grow. If a subse-
quent cooling rate is high enough, martensitic and/or
bainitic transformations from the austenite take place.
Materiali in tehnologije / Materials and technology 55 (2021) 1, 115–120 115
UDK 620.1:669.14.018.891:67.017 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 55(1)115(2021)
*Corresponding author's e-mail:
roman.celin@imt.si (Roman Celin)
The weld HAZ is narrow and difficult to investigate,
so a simulation of the welding thermal cycle is used to
produce a sample with a uniform CG HAZ micro-
structure that is suitable for investigations.
In most cases a weld thermal simulator and/or a
quenching dilatometer are used for thermal cycle simula-
tions, which are usually performed as a part of the
weldability investigation for the investigated steel.
9,10
Data acquired during thermal cycle simulations are used
to construct a continuous cooling transformation (CCT)
diagram, which is useful to monitor the microstructure
development in relation to the t
8/5
(cooling rate).
This study was performed with the use of a dilato-
meter. To conduct the continuous cooling transformation
tests of the coarse-grained HAZ (CG HAZ) for the ex-
perimental steel, several simulated thermal cycles were
applied. The simulation results were used to investigate
the correlation among the cooling rates, CG HAZ
microstructure, and hardness. The final goal was to de-
termine the t
8/5
cooling time interval for the experimental
steel-welding procedure.
2 EXPERIMENTAL PART
A steel characterization was carried out through
metallographic analyses and hardness measurements. A
12-mm plate made from the experimental laboratory
steel that fulfilled the S1100Q steel designation require-
ments was tested. The experimental steel has the follow-
ing mechanical properties: yield strength R
e
= 1186 MPa,
tensile strength R
m
= 1348 MPa, and an average hardness
HV5 = 446. Figure 1 shows the tempered martensite
microstructure of the experimental steel S1100 QT.
Cylindrical samples that were 4 mm in diameter and
10 mm in length were made for the welding thermal cy-
cle simulations in the dilatometer.
A TA Instruments DIL805A/D dilatometer was em-
ployed to simulate the weld CG HAZ in an argon atmo-
sphere. The samples were heated to the peak temperature
of 1350 °C with a heating rate of 200 °C/s, held at the
peak temperature for 3 s and then cooled by applying
five different time t
8/5
intervals ranging from5sto80s
(Table 1), the cooling curve was logarithmic to simulate
natural cooling during actual welding. The acquisition of
the cooling-cycle data was turned off at 200 °C. The
thermal-cycle parameters were selected based on previ-
ous work.
11
Table 1: Thermal-cycle simulation parameters
Heating rate
Peak tempera-
ture
Holding time
Cooling time
  t8/5
200 °C/s 1350 °C 3 s
5s,10s,20s,
40 s, 80 s
The dilatation data generated during the cooling were
recorded and the phase-transformation points were ob-
tained with the tangent method. Quantitative chemical
analysis of the experimental steel with a property of the
S1100 QL was made using ICP spectrometry. The chem-
ical composition in weight % is: 0.149 % C; 0.22 % Si;
0.95 % Mn; 0.003 % S; 0.52 % Cr; 0.38 % Cu; 1.31 %
Ni; 0.42 % Mo; 0.02 % V; 0.014 % Nb; 0.04 % N;
0.0006 % B; and 0.046 % Al.
Metallographic samples were prepared by grinding
and polishing, followed by chemical etching in a 2 %
nital solution. An optical microscope (OM) Microphot
FXA, Nikon and scanning electron microscope (SEM)
JEOL JSM-6500F with electron backscattered diffraction
(EBSD) were used for characterization of the simulated
CG HAZ microstructure.
The Vickers hardness of the samples from the ther-
mal cycle simulations was measured in accordance with
EN ISO 6507-1, applying a load of 5 kg (HV5). The re-
sults of the hardness testing were used to plot the "Time
t
8/5
vs. Hardness HV5" graph for the simulated CG HAZ
microstructure.
3 RESULTS AND DISCUSSION
3.1 Chemical composition
To assess the influence of the chemical composition
(carbon and alloying elements) on the steel’s weldability,
the term carbon equivalent CE is used. Equation (1) from
the standard EN 1011-2
12
is applicable to the investigated
experimental low-alloy high-strength steels with a car-
bon equivalent designation as CET:
CC
Mn Mo Cr Cu Ni
ET
=+
+
+
+
+
10 20 40
(1)
The carbon content of the experimental steel is
0.149 %, which ensures adequate weldability. However,
a CET value of a 0.364 suggests that preheat and control
of heat input during the welding process is needed to
avoid the occurrence of hydrogen-induced cracking (cold
cracking).
13
R. CELIN et al.: PROPERTIES OF THE SIMULATED COARSE-GRAINED MICROSTRUCTURE ...
116 Materiali in tehnologije / Materials and technology 55 (2021) 1, 115–120
Figure 1: Tempered martensite microstructure of the S1100 Q experi-
mental steel
3.2 Continuous cooling transformation diagram and
hardness
The dilatometry data that were recorded during the
welding thermal-cycle simulations were analyzed. The
heating rate and the first stage of cooling are the same
for all samples (Table 1). An increase in the temperature
causes linear thermal expansion in the sample. The trans-
formation points at the heating rate of 200 °C/s were
A
c1
= 754 °C and A
c3
= 850 °C. The temperatures A
c1
and
A
c3
are higher than the equilibrium temperatures A
e1
and
A
e3
because of the rapid heating rate.
14
During cooling, dilatation of the sample decreases
linearly from 1350 °C with the decrease in temperature
until it reaches the transformation start temperature. Due
to the low temperatures the diffusional transformations
are not possible, so the displacive transformations take
place (bainitic, martensitic). The dilatation again de-
creases linearly with cooling to room temperature. The
transformation start temperature (T
s
) and the transforma-
tion finish temperature (T
f
) for all the cooling rates were
determined from the plots (Table 2).
Table 2: Transformation start s and transformation finish temperature
t
8/5
cooling time (s) (cooling rate (°C/s))
5 (60) 10 (30) 20 (15) 40 (7.5) 80 (3.75)
Ts
(°C) 412 413 420 448 476
Tf
(°C) 265 300 340 345 360
The dilatation plots at the cooling stage of the simu-
lated thermal cycles t
8/5
= 5 s, 20 s and 80 s are shown in
Figure 2.
Based on the dilatometry data, the cooling diagram
for the simulated weld CG HAZ microstructure was con-
structed and is shown in Figure 3.
From Figure 3 it is clear that the transformation start
and transformation finish temperature rise gradually with
the increasing cooling time t
8/5
. The transformation at
continuous cooling occurs in a temperature range be-
tween 476 °C and 265 °C for all the applied cooling
rates.
R. CELIN et al.: PROPERTIES OF THE SIMULATED COARSE-GRAINED MICROSTRUCTURE ...
Materiali in tehnologije / Materials and technology 55 (2021) 1, 115–120 117
Figure 2: Dilatation vs. temperature cooling curves for the t
8/5
=5s,
20 s and 80 s
Figure 3: Cooling diagram of the simulated weld CG HAZ with T
s
and T
f
temperature
Figure 4: Optical micrographs of the simulated CG HAZ samples at
different t
8/5
(M – martensite; LB – lower bainite, UB – upper bainite)
Dependence of the simulated CG HAZ HV5 hardness
on the cooling time t
8/5
for the investigated experimental
steel is given in Figure 3. The average hardness of the
experimental S1100Q steel is 446 HV and is the result of
the quenched and tempered martensitic microstructure.
The application of a simulated thermal cycle with the
cooling times t
8/5
of 5 s, 10 s and 20 s softens the steel,
where the sample hardness drops to 424 HV5, 423 HV5,
and 421 HV5, respectively, all being below the hardness
of the base metal. As the cooling time t
8/5
increases to
80 s, the hardness further decreases to 334 HV5. The de-
cease of hardness can be attributed to the formation of
bainite.
3.3 Microstructure
A tempered martensite microstructure of the experi-
mental steel (base metal BM) is presented in Figure 1.
Figure 4 shows microstructure of the simulated CG
HAZ samples with cooling times t
8/5
of 5 s, 20 s and 80 s.
The microstructure consists of lower bainite and
martensite in the 5 samples with a consistently increas-
ing amount of upper bainite as the cooling times in-
crease. All the microstructures are coarse grained, as in-
tended by the thermal cycles, in comparison with
fine-grained unaffected base metal. The simulation peak
temperature of 1350 °C and a sufficient holding time en-
sure complete re-austenisation and rapid grain growth
that is typical for the CG HAZ.
The prior austenite grain boundaries are still notice-
able under the light microscope, but become more appar-
ent under the scanning electron microscope (SEM).
Figure 5a shows the tempered martensite micro-
structure of base metal (Figure 1). Figure 5b to and 5d
shows SEM images of microstructures corresponding to
the simulated samples. The carbide precipitates can be
clearly observed in the SEM images, the slow natural
cooling at lower temperatures allows for some self-tem-
pering of the martensite as small tempering carbides are
visible. The fine precipitates are located mainly inside
the martensite laths and in lower bainite in Figure 5b,
while the slightly larger carbides in Figure 5d belong to
the upper bainite.
In a lath martensitic structure, a prior austenite grain
is divided into packets of parallel laths, containing ex-
tended parallel blocks, and those blocks are further
sub-divided by laths, among which there are low-angle
R. CELIN et al.: PROPERTIES OF THE SIMULATED COARSE-GRAINED MICROSTRUCTURE ...
118 Materiali in tehnologije / Materials and technology 55 (2021) 1, 115–120
Figure 5: SEM images of the base metal and simulated CG HAZ at t
8/5
of5s,40sand80s
boundaries, while the packet and block boundaries are at
high angles.
15,16
Figure 6 shows band contrast images of the base
metal (Figure 6a) and the simulated HAZ at 1100 °C
with a slow cooling of 80 s (Figure 6b), as well as the
distribution of the grain-boundary misorientation of both
microstructures (Figure 6c). As the base metal shows
small equiaxed prior austenite grains (PAG) and a typical
martensitic sub-grain structure, the microstructure of the
simulated HAZ shows an evident increase of PAG, pack-
ets, and blocks, and correspondingly longer laths. Re-
garding the grain misorientation angles, the base metal
shows a lower ratio of low-angle boundaries (< 5°), an
almost constant share of boundaries with misorientation
angles between 5° and 45° with a slight increase between
10° and 20°, and an increased share of boundaries with
discrete misorientation angles of 55° and 60°. On the
other hand, the simulated HAZ microstructure has a
sharp increase of grain-misorientation angle of around 2°
(lath boundaries), a nearly linear decrease of share up to
22°, almost no grain boundaries with misorientation an-
gle between 22° and 50°, then similarly to the base
metal, and an increased share of boundaries with discrete
misorientation angles of 55° and 60°. The highest share
of boundary misorientation angles is at 60°, which repre-
sents the prior austenite, packet, and block boundaries.
4 CONCLUSIONS
The focus of this investigation was on the behaviour
of the experimental steel’s coarse-grained heat-affected
zone, which, in real welds, is desired to be as narrow as
possible.
The dilatometry analysis of the phase transformations
by cooling time was used as a first step of the weldability
assessment. The microstructure of the CGHAZ simu-
lated samples was evaluated with optical and a scanning
electron microscopy. The main conclusions regarding the
investigation of the experimental steel are:
Changes in the microstructure among different t
8/5
simulation samples were mainly an increase of lower and
upper bainite share and a decrease in the amount of
martensite.
The HV5 hardness of the simulated HAZ samples de-
creased by 112 HV5 from the base metal for the cooling
time t
8/5
of 80 s. The simulation cooling rate was slow
enough to induce the start of the re-tempering process,
which causes the hardness to drop.
The results of this investigation suggest that the t
8/5
of
the investigated experimental steel CG HAZ should be
between 5 s and 40 s. However this is preliminary assess-
ment based on the data available so far.
Further investigations of the experimental steel will
be carried out on a weld simulator, making it possible for
tensile test and Charpy notch toughness specimens to be
produced. The microstructure transformation of samples
in continuous cooling are in the temperature range be-
tween 476 °C and 265 °C, thus thermal simulation with
continuous cooling will be carried between 500 °C and
200 °C.
Acknowledgment
T h i sr e s e a r c hw a sm a d ea sap a r to f^ M R L J
(OP20.03532) research project co-financed by the Re-
public of Slovenia and the European Union under the
European Regional Development Fund.
R. CELIN et al.: PROPERTIES OF THE SIMULATED COARSE-GRAINED MICROSTRUCTURE ...
Materiali in tehnologije / Materials and technology 55 (2021) 1, 115–120 119
Figure 6: Band contrast images and distribution of grain boundary
misorientation
5 REFERENCES
1
J. Berneti~, B. Kosec, G. Kosec, Z. Burzi}, B. Podlipec, A. Nagode,
B. Karpe, S. Kanalec, F. Vodopivec, L. Kosec, Phenomena in pene-
trating piercing bullets in armored steel plate, Metalurgija, 55 (2016)
1, 95–98
2
O. Slyvinskyy, Y. Chvertko, S. Bisyk, Effect of welding heat input on
heat-affected zone softening in quenched and tempered armor steels,
High Temperature Material Processes: An International Quarterly of
High-Technology Plasma Processes, 23 (2019) 3, 239–253,
doi:10.1615/HighTempMatProc.2019031690
3
EN 10025-6:2004+A1 Hot rolled products of structural steels – Part
6: Technical delivery conditions for flat products of high yield
strength structural steels in the quenched and tempered condition
4
M. Lomozik, E. Turyk, Mechanical Properties of Welded Joints in
Steel S1100QL after Multiple Repair Welding, Biuletyn Instytutu
Spawalnictwa, (2018) 3, 7–15, doi:10.17729/ebis.2018.3/1
5
M. Kowalski, T. £agoda, F. ¯ok, V. Chmelko, Fatigue life of butt
weldments made of S1100QL steel, Journal of Machine Construction
and Maintenance, 113 (2019) 2, 7–13
6
P. Haslberger, S. Holly, W. Ernst, R. Schnitzer, Microstructure and
mechanical properties of high-strength steel welding consumables
with a minimum yield strength of 1100 MPa, J Mater Sci. 53 (2018)
9, 6968–6979, doi:10.1007/s10853-018-2042-9
7
M. Dun|er, I. Samard`i}, A. ]ori}, G. Salopek, Effects of real weld-
ing parameters of high-strength S1100QL steel on hardness and im-
pact energy properties, Metalurgija, 57 (2018) 4, 242–244
8
J. Górka, D. Janicki, M. Fidali, W. Jamrozik, Thermographic assess-
ment of the HAZ properties and structure of thermos-mechanically
treated steel, International Journal of Thermophysics, 38 (2017) 12,
doi:10.1007/s10765-017-2320-9
9
R. P. S. Sisodia, M. Gáspár, Physical simulation-based characteriza-
tion of HAZ properties in steels. Part 1. High-strength steels and
their hardness profiling, Strength of Materials, 51 (2019) 3, 490–499,
doi:10.1007/s11223-019-00094-5
10
M. S. Wêglowski, M. Zeman, M. £omozik, Physical Simulation of
Weldability of Weldox 1300 Steel, Materials Science Forum, 762
(2013), 551–555. doi:10.4028/www.scientific.net/MSF.762.551
11
R. Celin, J. Burja, G. Kosec, A comparison of as-welded and simu-
lated heat affected zone (HAZ) microstructures, Materials and tech-
nology, 50 (2016) 3, 455–460, doi:10.17222/mit.2016.006
12
EN 1011-2, Welding – Recommendations for welding of metallic
materials – Part 2: Arc welding of ferritic steels
13
T. Schaupp, W. Ernst, H. Spindler, T. Kannengiesser, Hydrogen-as-
sisted cracking of GMA welded 960 MPa grade high-strength steels,
International Journal of Hydrogen Energy, doi:10.1016/j.ijhydene.
2020.05.077
14
H. K. D. H. Bhadeshia, R.W. K. Honeycombe. Steels – Microstruc-
ture and Properties, 3th ed., Butterworth-Heinemann, Oxford 2006,
299
15
S. Morito, H. Tanaka, R. Konishi, T. Furuhara, T. Maki, The mor-
phology and crystallography of lath martensite in Fe-C alloys, Acta
Materialia, 51 (2003) 6, 1789–1799, doi:10.1016/S1359-6454(02)
00577-3
16
F. Zhen, K. Zhang, Z. Guo, J. Qu, Effect of martensite structure on
mechanical properties of an 1100 MPa grade ultra-high strength
steel, Journal of iron and steel research international, 22 (2015)7,
645–651, doi:10.1016/S1006-706X(15)30052-2
R. CELIN et al.: PROPERTIES OF THE SIMULATED COARSE-GRAINED MICROSTRUCTURE ...
120 Materiali in tehnologije / Materials and technology 55 (2021) 1, 115–120